A Field-Level Asset Mapping Dataset for England's Agricultural Sector

Area of Science:

• Agricultural Science
• Environmental Science
• Data Science

Background:

• The agriculture sector significantly contributes to greenhouse gas emissions.
• A lack of detailed farm-level data impedes effective transition finance and decarbonisation.
• Accurate asset-level information on ownership, land use, and production is missing.

Purpose of the Study:

• To develop an open-source, farm-level dataset to address data gaps in agriculture.
• To facilitate transition finance and decarbonisation efforts in the agricultural sector.
• To support financial instruments for sustainable agriculture.

Main Methods:

• Utilized natural language processing (NLP) and unsupervised learning techniques.
• Mapped farm names to spatial polygons to identify ownership and entity information.
• Extracted key attributes including addresses, land areas, crop types, production output, and geospatial coordinates.

Main Results:

• Successfully created a comprehensive farm-level dataset for England.
• Identified 117,116 distinct farming entities with detailed attributes.
• The dataset provides essential data for financial and environmental assessments.

Conclusions:

• Emerging farm-level datasets are critical for sustainable agriculture finance.
• This dataset can enhance the verification of carbon credits and sustainability-linked loans.
• Improved data quality will bolster risk assessment for climate finance in agriculture.